\begin{table}[!h]
\centering
\caption{Effect of Government Affiliation on Racism and Violence Framing\label{tab:embedreg}}
\centering
\resizebox{\ifdim\width>\linewidth\linewidth\else\width\fi}{!}{
\begin{threeparttable}
\begin{tabular}[t]{lllllllll}
\toprule
\multicolumn{1}{c}{\em{ }} & \multicolumn{8}{c}{\em{Similarity with Diaspora Dictionary (\%)}} \\
  & Racism (1) & Racism (2) & Racism (3) & Racism (4) & Violence (5) & Violence (6) & Violence (7) & Violence (8)\\
\midrule
Government & 6.28*** & 5.48*** & 6.29*** & 3.64*** & 5.11*** & 3.20*** & 4.98*** & 1.83**\\
 & (1.00) & (1.05) & (1.02) & (0.87) & (0.66) & (0.69) & (0.66) & (0.77)\\
ln Frequency 1 &  & -0.66 &  & -2.82*** &  & -1.08*** &  & -3.29***\\
 &  & (0.84) &  & (0.41) &  & (0.28) &  & (0.24)\\
ln Frequency 2 &  & -0.53* &  & -1.18*** &  & -2.15*** &  & -1.96***\\
 &  & (0.32) &  & (0.25) &  & (0.24) &  & (0.37)\\
\midrule
\addlinespace[]
\multicolumn{9}{l}{\textit{Statistics}}\\
\hspace{1em}Observations & 489 & 489 & 489 & 489 & 963 & 963 & 963 & 963\\
\addlinespace[]
\multicolumn{9}{l}{\textit{Fixed effects}}\\
\hspace{1em}Dictionary FE & No & No & Yes & Yes & No & No & Yes & Yes\\
\bottomrule
\end{tabular}
\begin{tablenotes}
\item Notes: $^{*}$p$<$0.1; $^{**}$p$<$0.05; $^{***}$p$<$0.01 White robust standard-errors are clustered on the attribute dictionary. Dependent variable is cosine similarity between respective object and attribute dictionaries. Controls for dictionary frequency are shown. Unit of analysis is the object-attribute word pair, which varies according to the object and attribute dictionary size.
\end{tablenotes}
\end{threeparttable}}
\end{table}
